Industrial expert system acquired by machine learning

Abstract This paper describes SERP, an expert system for repairing helicopter blades, Although very useful, it cannot evolve automatically with advances in technology and expert knowledge. To address this problem, we have designed an inductive learning system ( MEA) that uses rules or preclassified examples in order to incrementally build production rules. These rules are of the following form: if description of the fault & xrArr; repair and can be used by KEE ( knowledge engineering environment), an expert system generator. Both MEA and SERP use a representation based on structured objects. In this paper some aspects of MEA will be described, and the way MEA can extend andlor improve the iterative dichotomiier ( ID3) algorithm will be explained. MEA has been tested on a real-life application concerning the repairing of helicopter blades. Results show that it is efficient, simple to use, and helpful to the expert.